{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QAGM7VT3R4NM7LQ2MKCPO675IA","short_pith_number":"pith:QAGM7VT3","schema_version":"1.0","canonical_sha256":"800ccfd67b8f1acfae1a6284f77bfd402ac2df2bc476a43dbc813d7952791942","source":{"kind":"arxiv","id":"2106.04217","version":3},"attestation_state":"computed","paper":{"title":"Dynamic Sparse Training for Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Decebal Constantin Mocanu, Elena Mocanu, Ghada Sokar, Mykola Pechenizkiy, Peter Stone","submitted_at":"2021-06-08T09:57:20Z","abstract_excerpt":"Deep reinforcement learning (DRL) agents are trained through trial-and-error interactions with the environment. This leads to a long training time for dense neural networks to achieve good performance. Hence, prohibitive computation and memory resources are consumed. Recently, learning efficient DRL agents has received increasing attention. Yet, current methods focus on accelerating inference time. In this paper, we introduce for the first time a dynamic sparse training approach for deep reinforcement learning to accelerate the training process. The proposed approach trains a sparse neural net"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2106.04217","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-08T09:57:20Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c7cd4167673a167d10d7748bb8447edc5d25aaad6a88e5f1cbf93193a2b23a9a","abstract_canon_sha256":"0ce971ad05296e63b4fcdc21e9a7d21a903c3c8fffbb9308d02e026811efaaf3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:20:42.550143Z","signature_b64":"HLw1sLYVgqqtaw6NDSGzS00ae0xldZJ0s2uWJArkj4S0Kn+NddslmbygaQzJ/KIjdK0e58XEEGHLXY16jt1CCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"800ccfd67b8f1acfae1a6284f77bfd402ac2df2bc476a43dbc813d7952791942","last_reissued_at":"2026-07-05T04:20:42.549664Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:20:42.549664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamic Sparse Training for Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Decebal Constantin Mocanu, Elena Mocanu, Ghada Sokar, Mykola Pechenizkiy, Peter Stone","submitted_at":"2021-06-08T09:57:20Z","abstract_excerpt":"Deep reinforcement learning (DRL) agents are trained through trial-and-error interactions with the environment. This leads to a long training time for dense neural networks to achieve good performance. Hence, prohibitive computation and memory resources are consumed. Recently, learning efficient DRL agents has received increasing attention. Yet, current methods focus on accelerating inference time. In this paper, we introduce for the first time a dynamic sparse training approach for deep reinforcement learning to accelerate the training process. The proposed approach trains a sparse neural net"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.04217","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2106.04217/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2106.04217","created_at":"2026-07-05T04:20:42.549724+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.04217v3","created_at":"2026-07-05T04:20:42.549724+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.04217","created_at":"2026-07-05T04:20:42.549724+00:00"},{"alias_kind":"pith_short_12","alias_value":"QAGM7VT3R4NM","created_at":"2026-07-05T04:20:42.549724+00:00"},{"alias_kind":"pith_short_16","alias_value":"QAGM7VT3R4NM7LQ2","created_at":"2026-07-05T04:20:42.549724+00:00"},{"alias_kind":"pith_short_8","alias_value":"QAGM7VT3","created_at":"2026-07-05T04:20:42.549724+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.17909","citing_title":"NeuroTrails: Training with Dynamic Sparse Heads as the Key to Effective Ensembling","ref_index":63,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA","json":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA.json","graph_json":"https://pith.science/api/pith-number/QAGM7VT3R4NM7LQ2MKCPO675IA/graph.json","events_json":"https://pith.science/api/pith-number/QAGM7VT3R4NM7LQ2MKCPO675IA/events.json","paper":"https://pith.science/paper/QAGM7VT3"},"agent_actions":{"view_html":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA","download_json":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA.json","view_paper":"https://pith.science/paper/QAGM7VT3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.04217&json=true","fetch_graph":"https://pith.science/api/pith-number/QAGM7VT3R4NM7LQ2MKCPO675IA/graph.json","fetch_events":"https://pith.science/api/pith-number/QAGM7VT3R4NM7LQ2MKCPO675IA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA/action/storage_attestation","attest_author":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA/action/author_attestation","sign_citation":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA/action/citation_signature","submit_replication":"https://pith.science/pith/QAGM7VT3R4NM7LQ2MKCPO675IA/action/replication_record"}},"created_at":"2026-07-05T04:20:42.549724+00:00","updated_at":"2026-07-05T04:20:42.549724+00:00"}